In this application we propose a new hypothesis to explain the local Starling forces which regulate fluid balance between blood and tissue. The classical Starling forces for filtration and reabsorption are the global difference of hydrostatic and colloid osmotic pressure between plasma and tissue. In this renewal application we will evaluate the hypothesis that the local hydrostatic and colloid osmotic forces determine the Starling forces across the endothelial surface glycocalyx, the primary molecular filter, and not the entire wall. The hypothesis is to be tested in Specific Aim 1 is that the osmotic pressure difference across the glycocalyx is alrger than the blood-to- tissue osmotic pressure difference as the result of two processes: (1) sieving of the proteins in the glycocalyx; and (2) the reduction in the back diffusion of the tissue proteins into the space downstream of the glycocalyx by high water velocities through the breaks in the junctional strand. (Hypothesis 1). Tow hypotheses will be tested in Specific Aim 2: One states that changes in the size and frequency of the breaks in the junction strand resulting from changes in intracellular cAMP levels modify the osmotic pressure by changing the resistance to back diffusion of tissue protein (hence the effective concentration differences) not by changing the primary molecular filter. (Hypothesis 2) The second hypothesis to be tested is that a decrease in the thickness and/or organization of the endothelial cell glycocalyx caused by enzymatic degradation of the glycocalyx decreases the magnitude of the osmotic pressure difference exerted across the microvessel wall by directly modifying the primary molecular filter (Hypothesis 3). The hypothesis to be tested in Specific Aim 3 is that the junction associated molecule occludin is one of the key regulatory molecules determining the size and frequency of breaks in the junctional strands (Hypothesis 4). Individually perfused mammalian microvessels and mammalian endothelial cells in culture will be used to test these hypotheses. The design and interpretation of all experiments will be guided by a detailed 3 dimensional model of couple solute and water flows through the interendothelial cleft, taking into account the measured water flows across the microvessel wall actual ultrastructure of the junctional strands. This strategy has already been successfully used to test these ideas in frog mesenteric capillaries. The combined biophysical, ultrastructural, mathematical, and molecular approaches are expected to provide new understanding of the mechanisms which regulate fluid balance in normal tissue and after injury. The new information may lead to strategies to reduce edema formation and enhanced tissue recovery after injury and surgery.